Skip to content

Datawhale

Microduck-RL-4096x6000

Datawhale/Microduck-RL-4096x6000

Microduck-RL-4096x6000 is a reinforcement learning model from Datawhale. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for rsl-rl. The card lists the license as other.

这是 Every Embodied 为 Duck Together 共建计划完成的一次 Microduck 速度跟踪策略复现。训练基于 Pollen Robotics 的 microduckrl,使用 4096 个并行环境完成 6000 次 PPO 迭代,并保留中间 checkpoint、最终 ONNX、训练配置、TensorBoard 事件文件和连续闭环回放视频。

Downloads · 30 days

0

Access

Public

Updated Sep 4, 2026

Repo size

86.3 MB

Likes

0

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt63 MB · 73%

At a glance

Task
Reinforcement Learning
Library
rsl-rl
License
other
Access
Public
Created
Sep 4, 2026
Updated
Sep 4, 2026
SHA
c1cbc1a2
Task
Reinforcement Learning
Library
rsl-rl
License
other
Created
Sep 4, 2026
Updated
Sep 4, 2026